No subject

[Ferruh Yigit]

On 12/3/2021 6:18 AM, Christian Ehrhardt wrote:
> On Thu, Dec 2, 2021 at 3:30 PM Ferruh Yigit <ferruh.yigit at intel.com> wrote:
>>
>> On 11/30/2021 4:56 PM, christian.ehrhardt at canonical.com wrote:
>>> 5139502783  Ferruh Yigit     ethdev: fix typos
>>
>> These are simple 4 typo fixes, I think we can skip it. But if you think
>> we should get it, I can send a backport patch.
> 
> We clearly do not have any "hard need" for typo fixes.
> But if we do not apply too many of them sooner or later nothing will
> apply easily anymore.
> 
> So if it isn't feeling like a burden a backport would be appreciated
> to help the further lifetime of 19.11.x
> 

Sure, if it helps.

>> One of the commit that this patch fixes is after 19.11, that should be
>> reason why apply fails.
>>
>>> a1b2558cdb  Ferruh Yigit     kni: restrict bifurcated device support
>>
>> This doesn't apply to 19.11, OK to skip it.
> 
> ok
> 
>> Asking just to understand, the patch has fixes tag as:
>> Fixes: 631217c76135 ("kni: fix kernel deadlock with bifurcated device")
>>
>> The '631217c76135' is not exist in 19.11, isn't the script check
>> for this?
> 
> Not really, it is going the full chain of fixes, and if that is in the
> target release it is considered.
> So in this case it is:
> 
> 21.11 a1b2558cdb S F kni: restrict bifurcated device support (1.3.0r0
> (partially fixed in 21.05))
> 
> Which is saying
> 1. this is a1b2558cdb which fixes 631217c7
> 2. 631217c7 fixed something in 1.3.0r0
> 3. since you are >>19.11 you might want this
> 
> It does not (but probably could and sort out some false positives)
> check in #2 if that one is really applied on the stable branch.
> I think we could check the [ upstream commit ... ] for that and
> exclude or at least mark them in this case.
> 

ack, check is #2 can save some false positive